behavior characterization and development of lrfd resistance fact
TRANSCRIPT
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
1/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
2/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
3/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
4/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
5/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
6/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
7/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
8/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
9/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
10/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
11/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
12/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
13/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
14/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
15/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
16/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
17/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
18/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
19/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
20/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
21/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
22/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
23/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
24/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
25/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
26/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
27/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
28/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
29/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
30/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
31/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
32/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
33/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
34/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
35/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
36/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
37/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
38/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
39/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
40/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
41/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
42/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
43/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
44/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
45/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
46/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
47/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
48/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
49/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
50/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
51/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
52/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
53/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
54/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
55/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
56/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
57/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
58/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
59/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
60/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
61/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
62/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
63/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
64/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
65/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
66/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
67/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
68/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
69/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
70/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
71/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
72/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
73/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
74/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
75/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
76/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
77/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
78/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
79/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
80/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
81/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
82/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
83/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
84/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
85/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
86/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
87/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
88/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
89/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
90/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
91/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
92/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
93/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
94/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
95/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
96/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
97/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
98/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
99/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
100/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
101/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
102/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
103/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
104/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
105/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
106/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
107/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
108/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
109/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
110/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
111/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
112/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
113/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
114/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
115/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
116/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
117/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
118/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
119/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
120/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
121/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
122/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
123/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
124/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
125/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
126/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
127/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
128/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
129/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
130/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
131/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
132/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
133/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
134/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
135/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
136/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
137/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
138/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
139/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
140/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
141/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
142/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
143/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
144/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
145/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
146/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
147/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
148/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
149/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
150/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
151/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
152/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
153/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
154/235
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
155/235
138
full-scale vertical load tests conducted on steel H-piles installed in cohesive soils, the t-z
analysis using load-transfer curves and mBST, showed good agreement with measured load-
displacement response at the pile s head and load transfer along pile length. However,
conducting mBST in cohesionless soils is not easy, due to the tendency of borehole collapse
and/or the risk of losing the mBST shear head during testing. In addition, the mBST can only
capture the t-z curves required for calculating the skin-friction component of the pile
resistance, acceptable in the case of friction piles driven into mainly cohesive soil profiles,
but in the case of cohesionless soils, the end-bearing component cannot be neglected.
The main objective of this study is to utilize the t-z analysis method to accurately
predict the load-displacement response of vertically-loaded steel H-piles driven into
cohesionless soils. For this purpose, a modified Direct Shear Test (mDST) was developed tomeasure the t-z curves for different soil layers in the laboratory. In addition, the box for the
Direct Shear Test (DST) apparatus was used in a laboratory Pile Tip Resistance (PTR) test to
measure the q-w curve for the end-bearing soil layer. The proposed mDST and PTR
laboratory tests offer a simple and cost-effective method to directly measure the t-z and q-w
curves, respectively. These measured curves can be used with commercial software, TZPILE
v.2.0, developed by ENSOFT, Inc. to predict pile response using the t-z method. To validate
the proposed approach, two vertical static load tests (SLTs) were conducted on full-scale
instrumented steel H-piles driven into cohesionless soils at two different sites in Iowa and the
predicted responses were compared with those field measured.
5.3. Background
The use of DST to measure the soil-pile interface friction angle for steel piles was
initially proposed by Reddy et al. (2000) to estimate shaft capacity of steel piles driven into
sandy soils. Later, Pando et al. (2002) performed a series of tests to measure the fiber
reinforced polymer composite pile-soil interface friction angle. Pando et al. (2002) alsoconducted comparisons to the static shaft capacity, based on the Nordlund static method
(Nordlund, 1963) and concluded the static calculated capacity using the interface angle
measured, using DST, showed comparable values with the measured capacity. However,
none of the studies cited above used DST to measure the t-z curves required for the load-
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
156/235
139
transfer analysis, or used the measured curves to predict pile load-displacement response and
compared the analysis with field measured responses from pile static load test results.
For the end-bearing mechanism, numerous studies have investigated soil response
under the pile tip utilizing small-scale models or calibrated chambers; several have focused
on measuring the load-penetration response that represents the q-w curve at the pile s tip. For
example, Houlsby and Evans (1988) studied steel pile bearing capacity in layered silts and
sands, using a soil-pile laboratory model to determine the effect of different pile types on the
load-penetration response. Yasufuku and Hyde (1995) investigated the relationship between
the soil stress-strain properties at the pile s tip and the corresponding end-bearing capacity.
However, neither of the previous studies used the laboratory measured load-penetration curve
to conduct a load-transfer analysis or compared their results to measured responses fromlarge-scale SLTs. Furthermore, determining the minimum permissible size of a soil-pile
laboratory model with respect to the pile diameter or size (D) is not detailed in the
aforementioned studies and has no clear design standards.
Nevertheless, Bowles (1996) indicated that soil stresses and corresponding
displacements at a pile tip bearing on dense sand or sand-gravel deposits can be evaluated,
with reasonable accuracy, assuming a fictitious rigid footing placed on the bearing stratum.
To model a rigid footing in a test pit or a small soil box, the ASTM D 1194 standards
(ASTM, 2006) states that the soil box size should be at least four times the width of the
model footing. Given the depth of stresses influence below the pile tip ranges from 2D to 4D
in different soils and is approximately 1.5D in the case of coarse sand deposits (Houlsby and
Evans, 1988), the minimum required depth of the soil box should be at least 1.5D below the
model rigid footing, bearing on dense sand or sand-gravel deposits. The summary provided
above shows that for piles installed in dense sand or sand gravel deposits, a laboratory test
with model rigid footing placed in a soil box with a width greater than 4D and depth greater
than 1.5D can be used. However, when using small-scale rigid footings to represent pile tip
response in dense sand or sand gravel deposits, the measured q-w curves for the model
footing must be scaled or converted to represent the q-w of the large-scale pile as will be
discussed later.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
157/235
140
5.4. Field Testing
As part of an ongoing research project aimed at establishing LRFD resistance factors
for the design of deep foundations in the state of Iowa by accounting for local soil conditions,
ten vertical SLTs were conducted on full-scale instrumented steel HP 10x42 piles. The last
two SLTs were conducted on piles driven into cohesionless soil profiles (mainly sand) at Des
Moines County (test site ISU-9) and Cedar County (test site ISU-10). For ISU-9, the bridge
site was located southeast of Huron close to the Mississippi River. As for ISU-10, the bridge
was located south of Tipton at the intersection of Interstate-80 and the Iowa River. For both
test sites, a 15.8 m (52 ft) long test pile with 14.9 m (49 ft) embedded length was driven
between two anchor piles, 18.3m (60 ft) long with 16.4 m (54 ft) embedded length each. The
test piles were loaded, using a 2000 kN (440 kips) hydraulic jack, and the applied load was
measured using a 1300 kN (290 kips) load cell. T he Quick Test procedure outlined in the
ASTM D 1143 (ASTM, 2006) was used to load test the piles. In all cases, the load was
applied in 5% increments of the estimated nominal capacity and the load was kept relatively
constant at the end of each load increment for a duration ranging from 5 to 10 minutes until
the pile vertical displacement readings stabilized. After excessive vertical displacement at
approximately constant load was experienced, the piles were unloaded in five equal load
decrements. In addition to using four 25.4 cm (10 inch) displacement transducers to measurethe vertical displacement at the top of the two test piles, they were instrumented with strain
gauges along the shaft and near the tip.
Figures 5.1 and 5.2 show the locations of strain gauges installed on each side of the
web centerline of test piles ISU-9 and ISU-10, respectively. The locations of strain gauges
presented in the figures were determined, considering different soil layers at the two test sites
and included at least two gauges near the bottom of the piles to quantify tip resistance. The
soil profiles at the test sites were characterized, using in-situ SPT and CPT tests in addition to
several laboratory tests, including soil classification, Atterberg limits, as well as DST and
mDST. At ISU-10, a push-in pressure cell was used to monitor the lateral earth pressure
before and after driving of the piles as well as during SLT.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
158/235
141
5.4.1. Soil investigation
SPT and CPT results
The typical geological formation at ISU-9 consists of normally consolidated
Alluvium deposits of clay, sand, and gravel deposits. As shown in Figure 5.1, four different
soil layers were found during drilling. The first layer consisted of clay deposits extending to
4.8 m (15.7 ft) below the ground surface, followed by sand deposits up to 13.4 m (43.9
ft),underlain by a 2.6 m (8.5 ft) of granular material, and the bottom layer consisted of firm
silty clay material. The groundwater table at the time of in-situ testing was located at 5.2 m
(17.1 ft) below the ground s surface. Using laboratory tests conducted on soil samples
collected at ISU-9, the top soil layer was classified as low plasticity clay (CL) per the UnifiedSoil Classification (USCS), the second layer was classified as well-graded sand (SW), the
third as well-graded gravelly sand (SW), and the bottom layer as inorganic clay of high
plasticity clay (CH). In addition to the basic soil properties, Figure 5.1 summarizes the
measured CPT tip resistance (q c) and skin friction (f s) for ISU-9. Also included in Figure 5.1
are the SPT blow counts corrected for the effect of the soil s overburden pressure and the
corresponding relative density (D r %) after Terzaghi and Peck (1967).
The soil profile at ISU-10 was divided into three layers, where the first layer
consisted of sandy fill material extending to 4.6 m (15.1 ft) below the ground surface
followed by a 10.4 m (34.1 ft) layer of coarse sand with gravel as presented in Figure 5.2.
The bottom layer consisted of sand and boulders, and the groundwater table was located 3 m
(9.8 ft) below the ground surface at the time of in-situ testing. Using laboratory tests
conducted on the soil samples from ISU-10, the top soil layer was classified as well-graded
sand (SW) per the USCS, the second layer was classified as well-graded sand with gravel
seam (SW), and the bottom layer as well-graded gravelly sand (SW). As shown in Figure 5.2,
a 1.0 m (3 ft) layer of boulders was located almost in the middle of the second layer, which prevented the CPT from penetrating to the desired depth. Figure 5.2 summarizes the basic
soil properties, the measured q c and f s from the CPT results, the corrected SPT blow counts,
and the relative density (D r %) of different soil layers at ISU-10.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
159/235
142
5.4.2. Monitoring lateral earth pressure
In preparation for conducting conventional and modified laboratory DSTs at different
normal stresses to measure soil shear strength parameters and adequately capture the t-z andq-w curves along the soil-pile interface, expected ranges of lateral earth pressure ( h) acting
along the test piles were required to determine the appropriate normal stresses for use in the
DST and mDST tests. Figures 5.1 and 5.2 represent the location of DST/mDST for different
soil layers at ISU-9 and ISU-10. To complete this task, the angle of soil internal friction ()
and the over consolidation ratio (OCR) of soil layers were estimated, based on empirical
correlations with SPT and CPT in- situ tests. Based on SPT results, the and the OCR were
estimated after Peck, Hanson, and Thornburn (1974) and Mayne and Kemper (1988),
respectively. For CPT-based estimates, recommendations by Robertson and Campanella
(1983) and Mayne and Kulhawy (1990) were used. The at-rest lateral earth pressure
coefficient (K o) was then determined by using the equations recommended by Jaky (1944)
and Mayne and Kulhawy (1990) for Normally Consolidated (NC) and Over Consolidated
(OC) soils, respectively. Figures 5.1 and 5.2 summarize the K o, OCR, and the corresponding
effective v along the pile length for NC and OC soils for ISU-9 and ISU-10.
To validate the estimated lateral earth pressure calculated based on SPT and CPT
tests, a push-in pressure cell was installed at ISU-10 at 20 cm (8.0 inch) from the test pile
flange and 3.1 m (10 ft) below the ground surface. The pressure cell data were continuously
recorded during pile driving, re-striking, and SLT. On average, the effective lateral earth
pressure measured by the pressure cell at the time of SLT was 22 kPa (3.2 psi), very close to
the calculated value for NC soils at the same depth, using the theoretical correlations with
SPT and CPT results (see Figure 5.2).
5.4.3. Pile static load tests
For test pile ISU-9, the load distribution along the pile length was established and
presented in Figure 5.3, using the strain gauge data recorded at the end of each load
increment during SLT. Figure 5.3 shows the rate of load transferred in the pile during the last
nine load steps represented by the slope of the curves was almost zero within the top soil
layer, indicating the cohesive material did not significantly contribute to pile resistance.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
160/235
143
Within the second and third soil layers, an average rate of load transfer of 30 and 75 kN/m
(2.1 and 5.1 kips/ft) was observed, respectively. Furthermore, the load transferred to the
pile s tip during the test ranged from 0 to 210 kN (0 to 47.2 kips), indicating an end-bearing
of approximately 28% of the maximum applied load.
The measured total load-displacement response of the ISU-9 pile is presented in
Figure 5.4. Accordingly, pile response was approximately linear up to 521 kN (117.1 kips)
and sustained a maximum applied load of 754 kN (169.5 kips). Based on Davissons criterion
(Davisson, 1972), the pile nominal capacity was determined 704 kN (158.3 kips). The shaft
load-displacement was determined by integrating strain gauge data, using Simpsons
numerical integration rule (Figure 5.3). The load-displacement curve at the pile tip was then
back-calculated by subtracting the skin friction component from the total response measuredat the pile s head. A similar procedure was used for the test pile at ISU-10, although the
strain gauges were not functioning properly. This precluded an accurate separation of the
skin friction and the end-bearing components from the pile s total capacity. Figure 5.4
represents the load-displacement response measured at the pile s head for ISU-10, indicating
an approximate total nominal capacity of 580 kN (130.4 kips), based on Davissons criterion .
5.5. Direct Measurement of Load-Transfer Curves
5.5.1. Modified direct shear test to measure the t-z curves
Conventional DST was modified to directly measure the t-z curves at the soil-pile
interface for steel H-piles at ISU-9 and ISU-10. A square steel plate, with the same grade of
the steel piles, was fabricated to fit into the lower half of the DST mold with dimensions
equal to 100 mm (3.94 inch) in width and 40 mm (1.57 inch) in height. As shown in Figure
5.5, the steel plate is placed in the lower half of the shear box, while the upper half is filled
with soil material compacted to match the soil unit weight and estimated relative density
summarized in Table 5.1. Soil samples collected below the ground water table were
submerged in de-aired water during testing. A normal stress equal to the estimated horizontal
stresses on the surface of the pile at the depth of the tested sample was applied. The mDST
was conducted following the ASTM D3080-98 (ASTM, 2006). While performing the mDST,
the shear stress versus horizontal displacement was recorded after correction of the reduced
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
161/235
144
shear contact area, which represents the interface properties between the soil and the steel
plate, or the t-z curve at the soil-pile interface.
For the test pile at ISU-9, conventional and modified DSTs were conducted on soil
samples collected at depths of 2.6 m (8.5 ft), 7.0 m (23 ft), and 13.7 m (45 ft) below the
ground s surface, covering the main soil layers at the test site. As an example of the
measured properties, Figure 5.6 shows the Mohr-Coulomb failure envelope for the soil and
the soil-pile interface obtained from DST and mDST results, respectively, at 7.0 m (23 ft)
below the ground surface at ISU-9. As expected, the figure shows smaller values for
adhesion (a) and friction angle () at the soil-pile interface, when compared to soil shear
strength parameters. For all soil layers at ISU-9, Table 5.1 summarizes the values of the
water content, dry and total soil unit weights, relative density, and normal stresses usedduring laboratory testing, in addition to the measured soil shear strength parameters (i.e., c'
and '), and soil- pile interface properties (i.e., adhesion, a, and friction angle, ) , using the
DST and the mDST, respectively.
Figure 5.7 presents the measured t-z curves obtained using mDST at the soil-pile
interface compared to those back-calculated from the strain gauge readings obtained from the
SLT data at ISU-9. As can be observed from the figure, there is a good match between the
laboratory and field measured t-z curves, especially when comparing the maximum shear
stress values of the top and bottom soil layers, as the difference did not exceed 5%. The t-z
curves presented in Figure 5.7 were measured at normal stresses corresponding to the
measured and/or estimated lateral earth pressure in the field (see Table 5.1). For test site ISU-
10, DST and mDST tests were conducted following the same procedures used for ISU-9 at
different normal stresses as summarized in Table 5.1. The table also summarizes the soil
shear strength parameters and soil-pile interface properties measured for the test pile at ISU-
10. The test results for ISU-10 showed similar trends to those observed for ISU-9.
Using the load transfer analysis, the t-z curves were measured by utilizing mDST to
predict pile load-displacement response. However, to include the end-bearing component, the
q-w curve at the pile tip is needed. In the following section, another modification is proposed
to the DST apparatus to measure the q-w curve at the pile s tip.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
162/235
145
5.5.2. Pile tip resistance test to measure the q-w curves
As previously discussed, the response of the tip of a pile installed in sand deposits can
be modeled as rigid footing in the laboratory. A small-scale steel plate (representing afictitious rigid footing) was used to model the vertical load-displacement relationship ( q-w
curve) under the pile s tip for ISU-9 and ISU-10. The box with the DST apparatus,
dimensions of 100 mm (3.94 inch) in width and 60 mm (2.36 inch) in height (see Figure 5.8),
was used in a laboratory Pile Tip Resistance (PTR) test to measure the q-w curve for a small-
scale steel plate.
To confirm the suggested relative dimensions between the soil box and the model
plate summarized in the background section, steel plates with different sizes and shapes were
tested. The box width to plate width ratios of 4 and 5.7 and box depth to plate width ratios of
2.4 and 3.4 were used. These plates were embedded by 10 mm (0.40 inch) in the compacted
soil. For the model representing ISU-9, the soil sample collected at a depth of 15 m (49.2 ft)
(i.e., the depth of the pile tip) was compacted inside the box to match the estimated D r of
68%. For ISU-10, the sample collected at the pile s tip was compacted to achieve a D r % of
76%. For both PTR tests at ISU-9 and ISU-10, the loads were applied in controlled steps
equal to one-twentieth of the estimated maximum load, following testing procedures
provided by Yasufuku and Hyde (1995). For each loading increment, the corresponding
vertical displacement was measured until the rate of settlement was less than 0.01 mm/min.
When comparing the measured responses of different plates, the difference in the
load-displacement curves measured using both steel plate sizes was determined less than
10%. Furthermore, the effect of placing the plate on the soil s surface was evaluated and
found its effect is also less than 10%. Therefore, using different sizes and placement
locations of steel plates did not significantly affect PTR test results. Hence, the box s
dimensions relative to the used steel plates were considered acceptable.In addition to the previous laboratory checks on box dimensions, a finite element (FE)
analysis representing the used box in the PTR test, the soil, and the steel plate (with a box to
plate width ratio of 4.0) was conducted to confirm the dimensions of the box and the plate
have minimal effects on the measured responses. An axisymmetric FE model was used to
simulate the behavior, based on the Mohr-Coulomb constitutive model for the soil material,
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
163/235
146
using the commercial computer program, PLAXIS 7.2. In the axisymmetric analysis, the
model plate diameter of 28 mm (1.12 inch) was selected to provide the same cross-sectional
area of the actual steel plate. The plate was modeled as a non-porous linear-elastic material
with elastic modulus equal to 2x10 8 kPa (2.9x10 7 psi) and Poissons ratio equal to 0.2. The
soil s shear strength measurements from the DST laboratory test results (see Table 5.1) were
used for the required parameters of the Mohr-Coulomb soil constitutive model. The FE
model was manually adjusted to match the stress-displacement behavior measured from the
laboratory model. Presented in Figure 5.9 is the evaluated normal stress corresponding to the
prescribed vertical displacement of 7.0 mm (0.28 inch) for ISU-9. As seen from the figure,
the maximum stresses induced in the soil were concentrated directly below the steel plate and
degraded towards the bottom horizontal boundary to less than 18% of the maximum stress.Moreover, the maximum vertical displacement under the plate did not exceed 1.4D, where D
is the diameter of the steel plate, even less than the minimum required depth of the soil box
specified in the literature. Thus, based on experimental results and finite element analysis, the
dimensions of the used box in the PTR test are large, compared to plate dimensions, and
insignificantly affect the vertical load-displacement response of the load tested plates.
The load-displacement response of the PTR model plate with box to plate width ratio
of 4.0 was used in the t-z analyses for ISU-9 and ISU-10. The measured load-displacement
curves for the model plate were converted to load-penetration curves ( q-w curves) for the
full-scale pile at ISU-9 and ISU-10. According to Bowles (1988), the following expression
can be practically used to extrapolate the model plate loads in the case of cohesionless soils.
= The q pile and q model represent the applied load on the full-scale pile and the model
plate, respectively. A pile and A model represent the full-size and the model pile cross-sectional
areas, respectively. By using the cross-sectional area ratio in the previous equation, the
square shaped PTR model pile can be practically converted for the full-size actual pile (H-
shaped). For displacement ( w) calculations, Terzaghi and Peck (1967) provided an equation
to convert the measured tip displacement of the model pile to that of the full size pile in
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
164/235
147
cohesionless soils as follows:
=2
+
2
The w pile and w model represent the vertical displacement of the full-scale pile and the
model plate, respectively. In Figure 5.10, the q-w curves for the two test piles are presented
after extrapolation from laboratory model tests. As seen in Figure 5.10, both piles failed in
end-bearing at about the same maximum applied load of 52 kN (11.7 kips), corresponding to
an approximate vertical displacement of 5.08 mm (0.2 inch).
5.6. Load-transfer Analysis
In this section, the t-z and the q-w curves measured, utilizing the mDST and the PTR
tests, respectively, were used to model the skin-friction and the end-bearing components of
the test piles using the load-transfer analysis (or the t-z model). The responses of ISU-9 and
ISU-10 tests were predicted, using the t-z models and compared with measured responses.
5.6.1. Model description
In the t-z model, the pile is divided into several segments and is replaced by elastic
springs. The surrounding soil is represented as a set of non-linear springs, with one spring
depicting soil behavior at the pile s tip. The initial t-z model was developed by Coyle and
Reese (1966) and Suleiman and Coyle (1976) in cohesive and cohesionless soils,
respectively. Reddy et al. (1998) modified the model to account for pile elastic deformation
in computing the mobilized skin friction, using an elastoplastic soil-pile interface model. The
model was then improved to include a hyperbolic response for the springs representing the
soil-pile interface by Misra and Roberts (2006). In each t-z model developed for the tested
piles at ISU-9 and ISU-10, the HP 10x42 pile was divided into 50 segments with eachsegment represented by an elastic spring of constant stiffness term, AE=3.596x108 kN,
where E and A are the elastic modulus and the cross-sectional area of the pile, respectively.
To model the soil and soil-pile interface in the t-z models, the t-z and the q-w curves,
respectively, were measured by using the mDST and the PTR tests.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
165/235
148
5.6.2. Load-displacement curves
For ISU-9, the response of the full-scale test was evaluated using two different t-z
analyses. Initially, the analysis was conducted using only the t-z curves measured by utilizingthe mDST (i.e., assuming no end-bearing and only including shaft resistance in the analysis),
called TZ-S-mDST, to compare the predicted response with the shaft load-displacement
relationship calculated using strain gauge data. The analysis was then conducted, including
both the t-z and q-w curves (TZ-T-mDST) to compare the predicted response with the
measured total load-displacement relationship (i.e., measured load-displacement at the pile
head). Figure 5.11 shows the predicted shaft and total load-displacement relationships
compared with the measured responses. As shown in Figure 5.11, the predicted shaft load-
displacement curve, using the TZ-S-mDST model, perfectly matches the measured
relationship (with a difference of only 3.3% on the conservative side). On the other hand, the
predicted total load-displacement relationship also provides a very good match of the
measured total response along the first portion of the curve and a difference of 4% in the
nominal capacity at the plastic portion, based on Davissons criterion (see Table 5.2). In
addition to comparing load-displacement responses, Figure 5.12 represents the load
distribution along pile length at an applied load of 754 kN (170 kips), which represents the
maximum applied load in the field, calculated by using the TZ-T-mDST model and
compared to the measured pile response. This figure shows TZ-T-mDST provides a very
good prediction of the load distribution along pile length compared to the loads based on
strain-gauge measurements. Hence, the t-z analysis succeeded to accurately capture the load-
displacement response and load transfer along pile length for ISU-9.
The total load-displacement response of the test pile at ISU-10 was predicted using
the t-z model based on the t-z and q-w curves attained from the mDST and PTR tests,
respectively (see Figure 5.13). As shown in the figure, the slope of the first portion of the predicted load-displacement response is about 12% stiffer than the measured response.
Furthermore, the capacity estimated using Davissons criterion from the TZ -T-mDST
analysis was about only 4% higher than the pile capacity estimated from the measured
response (see Table 5.2). Therefore, the TZ-T-mDST analysis matches the measured load-
displacement response for ISU-10 with acceptable accuracy. Based on comparisons with
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
166/235
149
measured responses for ISU-9 and ISU-10, it is evident the load-transfer analysis, based on
laboratory measured t-z and q-w curves, respectively, using the mDST and PTR tests,
provided an accurate and effective means of predicting the load-displacement response of
vertically-loaded steel piles in sandy soils.
5.7. Summary and Conclusions
The main goal of this study was to improve the prediction of the load-displacement
response of steel H-piles driven into cohesionless soils. The t-z analysis was used for this
purpose, utilizing TZPILE v.2.0 software. A laboratory modified Direct Shear Test (mDST)
and Pile Tip Resistance (PTR) test were proposed to improve t-z analysis by measuring load
transfer curves. The mDST and PTR basically provided a measure for the t-z and q-w curves,respectively, required for t-z analysis and represent the skin-friction and the end-bearing of
steel piles driven into sand soils. As part of this study, two static load tests were conducted
on full-scale instrumented steel H-piles at two locations (ISU-9 and ISU-10) with
cohesionless soil profiles. The soil investigation program for both test sites included in-situ
tests, such as SPT, CPT, and push-in-pressure-cells, as well as laboratory tests, such as soil
classification and DST. The t-z analysis, based on mDST and PTR measurements, provided
the shaft load-displacement response (TZ-S-mDST), as well as the total load-displacement
response (TZ-T-mDST). The results were compared to the responses measured in the field. A
summary of the major findings is presented below.
The laboratory measured t-z curves obtained, using mDST, were compared to
those measured in the field using strain gauges readings. It was found the
differences between them at the maximum shear stress did not exceed 5%.
The PTR test was conducted to measure the q-w curve, using different sizes and
placements of the small-scale steel plate. It was determined the effect of changing
these factors on the results is minimal and did not exceed 10%. A FE model was utilized to assess the effect of the PTR box boundaries on the
small-scale steel plate behavior, where the displacements induced in the model
soil did not exceed 1.4D. Hence, the PTR box dimensions are acceptable.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
167/235
150
For the test pile at ISU-9, the predicted load-displacement relationship for the
shaft using the TZ-S-mDST model perfectly matches the measured response.
When adding the end-bearing component to the ISU-9 analysis (i.e., TZ-T-
mDST), the model provided a very good match of the measured load distribution
as a function of depth, as well as the total load-displacement curve, with only
3.3% softer response along the elastic portion of the curve and 4% lower capacity
at the plastic portion.
The TZ-T-mDST model also showed very good prediction of the measured total
load-displacement response for ISU-10 with an acceptable error of about 12%
along the elastic portion of the curve and about 4% at the plastic portion.
Based on overall response predictions for the two test piles, the load transferanalysis, with t-z and q-w curves measured using the mDST and PTR tests,
respectively, has proven to provide a very good match of the measured responses,
which offer a simple and cost-effective procedure to predict steel pile responses.
Finally, similar studies can be conducted for other pile types, such as concrete piles,
and the proposed procedure can help incorporate serviceability limits into the LRFD design
of deep foundation, based on predicting pile load-displacement response.
5.8. Acknowledgments
The study reported in this paper was conducted by the authors as part of the ongoing
research project TR-573: Development of LRFD Design Procedures for Bridge Pile
Foundations in Iowa funded by the Iowa Highway Research Board (IHRB). The authors
express their gratitude to the IHRB and members of the project Technical Advisory
Committee for their guidance and advice. The members of this committee represent Office of
Bridges and Structures, Soils, Design Section, and Office of Construction of the Iowa DOT,
FHWA Iowa Division, and Iowa County Engineers. Finally, special thanks to Kam Weng Ng
and Matthew Roling for their help during the field/laboratory tests.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
168/235
151
5.9. References
AbdelSalam, S. S., Suleiman, M. T. , and Sritharan, S. (2010). Modeling Axially-loaded
Friction Steel H-Piles using the Load-Transfer Approach Based on a Modified
Borehole Shear Test . Submitted to Geotechnical Testing Journal , ASTM.
Alawneh, A. S. (2006). Modeling Load Displacement Response of Driven Piles in
Cohesionless Soils under Tensile Loading . Computers and Geotechnics , 32.
ASTM (2006) . Annual Book of ASTM Standards, Soil and Rock (I). American Society for
Testing and Materials , Vol. 4.08, ASTM, Philadelphia, PA.
Bowles, J. (1988). Foundation Analysis and Design. 4 th Edition, McGraw-Hill, New York.
Coyle, H. M. and Reese, L. C. (1966). Load Transfer for Axially-loaded Piles in Clay .
Journal of Soil Mechanics and Foundation Engineering Division , 92(2), 1- 26.
Davisson, M. (1972) . High Capacity Piles. Proceeding: Soil Mechanics Lecture Series on
Innovations in Foundation Construction , ASCE, IL Section, Chicago, IL, pp. 81 112.
De Gennaro, V., Said, I., and Frank, R. (2006). Axisymmetric and 3D Analysis of Pile Test
using FEM . Numerical Methods in Geotechnical Engineering , 2006 Taylor &
Francis Group, London, ISBN 0-4 15-40822-
El-Mossallamy, Y. (1999). Load -Settlement Behavior of Large Diameter Bored Piles in
Over-consolidated Clay . Proceeding: The 7th International Symposium on Numerical Models in Geotechnical Engineering , September 1999, 433-450.
Rotterdam, Balkema.
Engin, H. K., Septanika, E. G., and Brinkgreve, R. B. (2007). Improved Embedded Beam
Elements for the Modeling of Piles . Proceeding: The 10 th International Symposium
on Numerical Models in Geotechnical Eng. NUMOG X, Rhodes (Greece).
Houlsby, G. T., Evans, K. M., and Sweeney, M. (1988). End bearing capacity of model piles
in layered carbonate soils . Proc. Int. Conf. on Calcareous Sediments , Perth,
Australia, Balkema, Vol. 1, 209 214.
Jaky, J. (1944). The Coeffici ent of Earth Pressure at Rest . Journal of Society of Hungarian
Architects and Engineers , Budapest, Oct. 1944, pp. 355-358.
Mayne, P. W. and Kemper, J. B. ( 1988). Profiling OCR in Stiff Clays by CPT and SPT .
Geotechnical Testing Journal , ASTM, Vol. 11, No. 2, pp. 139-147.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
169/235
152
Mayne, P. W. and Kulhawy , F. H. (1990). Manual on Estimating Soil Properties for
Foundation Design . Rpt. EL-6800, Electric Power Research Inst., Palo Alto, 306 p.
Misra, A. and Chen, C. H. (2004). Analytical Solutions for Micropile Design under Tension
and Compression . Geotechnical and Geological Engineering , 22(2), 199-225.
Misra, A. and Roberts, L. A. ( 2006). Probabilistic Analysis of Drilled Shaft Service Limit
State Using the t -z Method. Canadian Geotechnical Journal , 1324-1332 (2006).
Nordlund, R. L. (1963). Bearing capacity of piles in cohesionless soils . Journal of Soil
Mechanics and Foundation Engineering , JSMFE, Vol. 89, No. SM 3.
Pando, M., Filz, G., D ove, J., and Hoppe, E. (2002). Interface Shear Tests on FRP
Composite Piles . ASCE International Deep Foundations Congress , Orlando, FL,
Vol.2, 1486-1500.Peck, R. B., Hanson, W. E., and Thornburn, T. H. (1974). Foundation Engineering .John
Wiley & Sons.
Reddy, E. S., Chapman, D. N., and Sastry, V. V. (2000). Direct Shear Interface Test for
Shaft Capacity of Piles in Sand . Geotechnical Testing Journal , Vol. 23, No. 2.
Reddy, E. S ., OReilly M., and Chapman, D. A. (1998). Modified t -z Model - A Software
for Tension Piles . Computers and Structures , 68: 613 25.
Roberts, L. A., Gardner, B. S., and Misra, A. (2008). Multiple Resistance Factor
Methodology for Service Limit State Design of Deep Foundations using a t -z Model
Approach . Proceeding: The Geo-Congress 2008, New Orleans, LA .
Robertson, P. K. and Campanella, R. G. (1983). Interpretation of Cone Penetration Tests .
Canadian Geotechnical Journal , Vol. 20, No. 4, Nov. 1983.
Schmertmann, J. B. (1978) .Guidelines for Cone Penetration Test, Performance and Design .
Report No. FHWA-TS-78-209, U.S. DOT , Washington, D.C., pp. 145.
Suleiman, I. B. and Coyle, H. B. (1976). Uplift Resistance of Piles in Sand. Journal of
Geotechnical Engineering , ASCE 1976; 102(GT5):559 62.
Terzaghi, K. and Peck, R. B. (1967). Soil Mechanics i n Engineering Practice . John Wiley
& Sons.
Yasufuku, N. and Hyde, A. F. (1995). Pile End -bearing Capacity in Crushable Sands .
Geotechnique 45, No.4, pp. 663-676.
http://www.cpt-robertson.com/pdfs/Interpretationofcptrobcamp.pdfhttp://www.cpt-robertson.com/pdfs/Interpretationofcptrobcamp.pdfhttp://www.cpt-robertson.com/pdfs/Interpretationofcptrobcamp.pdfhttp://www.cpt-robertson.com/pdfs/Interpretationofcptrobcamp.pdf -
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
170/235
153
Table 5.1: Soil and soil-pile interface shear strength parameters measured using the SPT, CPT,DST, and mDST at different depths
Test
SiteDepth(m) w c%
dry kN/m 3
bulkkN/m 3 D r %
Normal StressRange DST mDST
P n(kPa)
c'(kPa) '
a(kPa)
ISU-9
2.6 24% 18.8 23.4 163 10 20 * 30 N/A N/A 9.7 14.2
7.0 23% 18.5 22.9 38% 60 100 * 128 0 37.0 1.6 25.5
13.7 18% 21.4 25.2 68% 108 206 * 295 0 39.1 0 28.2
ISU-10
2.1 15% 17.6 20.2 40% 25 * 40 60 7.2 37.6 3.6 28.7
7.6 18% 20.3 24.0 70% 60 * 120 170 32.2 29.6 0 31.5
13.7 7% 22.2 23.7 76% 120 * 220 310 0.2 41.3 0 33.5For unit conversion: 1 m = 39.37 inch; 1 kN/m 3 = 6.24 lb/ft 3; and 1 kPa = 0.145 psi. Calculated based on SPT data after Terzaghi and Peck (1967); * Represents the normal stress correspondingto measured lateral earth pressure and selected for TZ-mDST analyses; and Value of the undrained shearstrength in kPa for the clay layer calculated based on CPT test results using correlations from Schmertmann etal. (1978), assuming N k =15.
Table 5.2: Summary of the t-z model findings compared to the field test results
TestSite Model
Model DescriptionLoad-Displacement
Curve % Difference to SLT*
t-z curves (skin-
friction)
q-w curve (end-
bearing)
Slope of FirstPortion
(kN/m)
DavissonCapacity
(kN)
Slope ofFirst
Portion
DavissonCapacity
ISU-9
SLT Strain Gauges Readings 132,178 704 Measured Pile Response
TZ-S-mDST mDST Ignored 127,779 550 3.3% softer 22%lowerTZ-T-mDST mDST PTR 127,779 675 3.3% softer 4% lower
ISU-10
SLT Strain Gauges Readings 87,391 580 Measured Pile Response
TZ-S-mDST mDST Ignored 98,036 484 12% stiffer 17%lowerTZ-T-mDST mDST PTR 98,036 603 12% stiffer 4% higher
For unit conversion: 1 m = 39.37 inch; and 1 kN = 0.224 kips.*Difference between the load-displacement curves adapted from the t-z analyses and the pile measuredresponse during SLT.
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
171/235
154
Figure 5.1: Summary of soil tests conducted at ISU-9 including: tip resistance (q c) and skinfriction (f s) from CPT;SPT N-values and corresponding D r %;soil classification; calculated K o
and OCR for NC and OC soils; corresponding effective lateral earth pressure; depths of
DST/mDST; and strain gauge locations along the test pile
0 20 40 60 80
0
12
3
4
5
6
7
8
9
10
11
12
1314
15
16
17
18
19
0 100 200 300
qc (kPa) x1000
D e p t h
( m )
fs (kPa)
fsqc
GWT
Low Plastic Clay (CL)d = 18.8 kPa
Well-graded Sand(SW)
= 22.9kPa
Well-graded Sandw/gravel (SW)
= 25.2 kPa
High Plastic Clay (CH) = 23.8 kPa
GroundSurface
CPT Results
0 25 50 75 100
0 10 20 30 40
Dr%
SPT N-values
SPTCorr. SPTDr % (SPT)
SG: StrainGauge Location
SG-1
SG-2
SG-3
SG-4
SG-5
SG-6SG-7
PileTip
Pile Head
0 1 2 3 4 5 6
0.0 0.4 0.8 1.2
OCR
Ko value
Ko SPT-NCKo SPT-OCKo CPT-NCOCR-SPT
DST - 3Dr%= 68%
DST - 2Dr%=38%
DST - 1
DST: DirectShear TestLocation
0 1 2 3 4 5
0 100 200 300
0
5
10
15
20
25
30
35
40
45
50
55
60
' h (kPa) x10
e p t h
( f t )
' h (psi)
SPT-NCSPT-OCCPT-NC
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
172/235
155
Figure 5.2: Summary of soil tests conducted at ISU-10 including: tip resistance (q c) and skinfriction (f s) from CPT;SPT N-values and corresponding D r %;soil classification; calculated K o
and OCR for NC and OC soils; corresponding effective lateral earth pressure; depths ofDST/mDST; and strain gauge locations along the test pile
0 10 20 30
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
0 50 100 150 200
qc (kPa) x1000
D e p t h
( m )
fs (kPa)
fs qc
GWT
Well-graded Sand(SW)
= 17.6 kPa
Boulders
Well graded Sand w/Gravel (SW)
= 23.7 kN/m 3
Well-graded GravellySand (SW)
= 2 3.1 kN/m 3
Well-graded Sandw/Gravel (SW)= 24.0 kN/m 3
CPT Results
GroundSurface
0 25 50 75 100
0 25 50 75 100
Dr%
SPT N-values
SPTCorr. SPTDr % (SPT)
SG-2
SG-3
SG-4
SG-5
SG-6SG-7
SG-8PileTip
SG: StrainGauge Location
Pile Head
0 1 2 3 4 5 6
0.0 0.5 1.0 1.5 2.0
OCR
Ko value
Ko SPT-NCKo SPT-OCKo CPT-NCOCR-SPT
DST - 3Dr%= 76%
DST - 2Dr%=70%
DST - 1Dr%=40%
DST: Direct ShearTest Location
0 10 20 30 40 50
0 100 200 300
0
5
10
15
20
25
30
35
40
45
50
55
60
' h (psi)
D e p t h
( f t )
' h (kPa)
SPT-NC
SPT-OCCPTP.Cell
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
173/235
156
Figure 5.3: Load distribution along the pile length calculated from measured strains at differentapplied loads for test site ISU-9
Figure 5.4: Measured load-displacement response at the pile head during SLT for ISU-9 andISU-10, and separated skin-friction and end-bearing components for ISU-9
0 25 50 75 100 125 150 175 200 225
0
10
20
30
40
50
0
2
4
6
8
10
12
14
16
0 200 400 600 800 1000
Load (kips)
e p t h
( f t )
D e p t h
( m )
Load (kN)
42131220263353438521607678754
Load (kN)
Ground
Low Plastic Clay
Well-graded Sand w/Gravel
28% end-bearing
Well-gradedSand
0 20 40 60 80 100 120 140 160 180
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0
2
4
6
8
10
12
14
16
18
20
0 100 200 300 400 500 600 700 800
Load (kips)
D i s p
l a c e
e n t
( i n c h
)
D i s p
l a c e m e n t
( m m
)
Load (kN)
Total Load (ISU-9)Skin-friction (ISU-9)
End-bearing (ISU-9)Total Load (ISU-10)
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
174/235
157
Figure 5.5: Overview of the mDST unit as used to measure the t-z curves
Figure 5.6: Mohr-Coulomb failure envelope for the soil and the soil-pile interface obtainedusing DST and mDST, respectively, at depth of 7 m forISU-9
P normal
P shear
120 mm100 mm
Vertical disp.
Dial Gauge
Horizontal disp.Dial Gauge
Shear StressDial Gauge
Steel Plate
Sand
Normal StressDial Gauge
20 mm
20 mm
PressureLines
PressureLines
Porous Plate
Porous Plate
Water Level
Loading Frame
Loading Plate
Shear BoxUpper Half
Shear BoxLower Half
Shear Box Container
y = 0.754xR = 0.932
y = 0.476x + 1.665R = 0.975
0 5 10 15 20 25 30
0
5
10
15
20
25
0
20
40
60
80
100
120
0 50 100 150 200
Normal Pressure (psi)
S h e a r S t r e s s
( p s i
)
S h e a r S t r e s s
( k P a
)
Normal Pressure (kPa)
DST
mDSTc = 0 kPa, = 37o
a = 1.6 kPa, = 25.5o
Well-gradedSand (SW)
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
175/235
158
Figure 5.7: Comparison of t-z curves developed from mDST with those back-calculated fromstrain gauges(SG)for the soil layers at test site ISU-9
Figure 5.8: Overview of the PTR unit as used to measure the q-w curves
0 0.05 0.1 0.15 0.2 0.25 0.3
0
2
4
6
8
10
12
14
16
0
20
40
60
80
100
120
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Displacement (inch)
n i t S h e a r S t r e s s
( p s i )
U n i t S
h e a r S t r e s s
( k P a
)
Displacement (mm)
SG 14.3m (46.9ft) mDST 13.7m (44.9ft)
SG 7.6m (24.9ft) mDST 7m (22.9ft)
SG 2.3m (7.5ft) mDST 2.5m (8.2ft)
Sand
ModelPile
60 mm
20 mm
10 mm
VerticalDial Gauge
Normal StressDial GaugePressure
Lines
Steel BoltsWater Level
Added 20 mm HeightSpacer Plate
Loading Frame
Shear BoxUpper Half
Shear Box Container
P normal
120 mm100 mm
25 mm
Shear BoxLower Half
Steelplate
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
176/235
159
Figure 5.9: FE model results representing the PTR test conducted for ISU-9
Figure 5.10: Comparison of the q-w curves obtained for the two test piles after extrapolationfrom the PTR test results
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
0
2
4
6
8
10
12
0
10
20
30
40
50
60
0 1 2 3 4 5 6 7 8 9 10 11
w (inch)
q ( k i p s )
q ( k N
)
w (mm)
q-w curve for ISU-9q-w curve for ISU-10
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
177/235
160
Figure 5.11: Load-displacement responses based on different t-z analyses and compared withmeasured response for test site ISU-9
Figure 5.12: Load distribution along the pile length calculated using TZ-T-mDST andcompared with measured values at test site ISU-9
Figure 5.13: Load-displacement responses calculated using TZ-T-mDST and compared withmeasured values at test site ISU-10
0 20 40 60 80 100 120 140 160 180
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0
24
68
10
12
14
1618
20
0 100 200 300 400 500 600 700 800
Load (kips)
D i s p
l a c e
e n t
( i n c
h )
D i s p
l a c e m e n t
( m m
)
Load (kN)
SLT (Shaft)TZ-S-mDST (Shaft)SLT (Total)TZ-T-mDST (Total)
0 25 50 75 100 125 150 175 200 225
0
10
20
30
40
50
0
2
4
6
8
10
12
14
16
0 200 400 600 800 1000
Load (kips)
D e p t h
( f t )
D e p t h
( m )
Load (kN)
Strain GaugesTZ-T-mDST
Ground
Low Plastic Clay
Well-graded Sand w/Gravel
Well-gradedSand
0 20 40 60 80 100 120 140 160
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
024
68
10
1214161820
0 100 200 300 400 500 600 700
Load (kips)
D i s p l a c e
e n t
( i n c h
)
D i s p l a c e m e n t
( m m
)
Load (kN)
SLT (Total)TZ-T-mDST (Total)
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
178/235
161
CHAPTER 6: PILE RESPONSE CHARACTERIZATION USING A FINITEELEMENT APPROACH
Sherif S. AbdelSalam 1; Sri Sritharan, M. ASCE 2; Muhannad T. Suleiman, M. ASCE 3
A paper to be submitted to a conference proceeding
6.1. Abstract
The study presented herein focused on improving the accuracy and efficiency of
simplified two-dimensional finite element (FE) models in simulating the response of
vertically-loaded steel piles driven into cohesive and cohesionless soils. For this purpose,
data from four static load tests (SLTs) conducted on instrumented steel H-piles driven intoclay and sand soil profiles accompanied by various in-situ and laboratory soil tests were
used. The behavior of the load-tested piles was simulated using axisymmetric FE analyses,
based on the Mohr-Coulomb soil constitutive model by means of PLAXIS 7.2. Following
identification that the elastic modulus of cohesive soils and the soil-pile interface properties
are poorly represented in the analysis, these capabilities were improved first. The modulus of
clay soil was obtained from a Consolidated Undrained (CU) Triaxial soil laboratory test,
while the modified Direct Shear Test (mDST) was utilized to provide an accurate laboratory
measurement of the soil-pile interface properties. With these refinements, the FE model
provided an improved prediction for the load-displacement response of the four test piles
without using sophisticated soil constitutive models that require conducting costly and time
consuming tests.
Keywords: Finite Element; Pile Foundations; Soil-structure Interaction; Load-displacement;Interface Element; Soil Elastic Modulus.
1 Graduate Student and Research Assistant, Dept. of Civil, Construction, and EnvironmentalEngineering, Iowa State University, 355 Town Engineering, Ames, IA 50011-3232, E-mail :[email protected]
2 Wilson Engineering Professor and Associate Chair, Dept. of Civil, Construction and EnvironmentalEngineering, Iowa State University, 376 Town Engineering, Ames, IA 50011-3232, E-mail: [email protected]
3 Assistant Professor, Dept. of Civil and Environmental Engineering, Lehigh University, STEPsBuilding, Room 326, 1 West Packer Avenue, Bethlehem, PA, 18015. E-mail: [email protected]
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected] -
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
179/235
162
6.2. Introduction
An analysis approach to predict the load-displacement response of axially-loaded
piles can facilitate incorporation of both strength and serviceability limit states in routine
design of deep foundations. According to Guo and Randolph (1998), analytical methods used
to characterize pile behavior can be classified into three major categories: (1) approximate
closed-form solutions, (2) one-dimensional numerical algorithms, and (3) boundary or finite
element methods. The first two categories depend on load-transfer analysis, an iterative
technique for solving the nonlinear differential equations of load-transfer using the finite-
difference method (Coyle and Reese, 1966; Suleiman and Coyle, 1976). The third category
depends on a more comprehensive modeling technique and accurate characterization of
material behavior.
Recently, numerous studies have successfully adapted the load-transfer analysis (t-z
analysis) to model the pile behavior under axial loads (Misra and Chen, 2004; Misra and
Roberts, 2006; Roberts et al., 2008; AbdelSalam et al., 2010). Although the t-z analysis
provides the load-displacement response at the pile head and load transfer as a function of
pile depth, it does not provide the stresses and strains induced in the surrounding soil
medium. Consequently, the t-z method may not provide insight into the response of soil and
pore water pressure development around the pile (Wathugala and Desai, 1989). Therepercussion of not having this knowledge is that the influence of the load transferred
through one pile on adjacent piles in a pile group or nearby shallow or deep foundations may
not be known. In these cases, it is important to use a detailed model to examine the load
transfer in a pile. For vertically-loaded piles, the soil properties and the interface element
behavior are the most important parameters that control the accuracy of the finite element
(FE) analysis outcomes. More specifically, De-Gennaro et al. (2006) and Engin et al. (2007)
reported soil-pile interface properties and the soil modulus typically estimated, based on
empirical correlations, largely dictate FE analysis results. Therefore, utilizing more detailed
laboratory tests to obtain the soil constitutive properties may be appropriate (De-Gennaro et
al., 2006; Engin et al., 2007). The accuracy of the FE analysis results is also dependent on the
pile material and construction techniques (El-Mossallamy, 1999).
The objective of this study is to examine the accuracy and efficiency of a simplified
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
180/235
163
Mohr-Coulomb soil constitutive model in simulating the response of vertically-loaded steel
H-piles driven into cohesive and cohesionless soils by utilizing a two-dimensional
axisymmetric FE analysis using PLAXIS 7.2. For this purpose, data from four vertical static
load tests (SLTs) conducted on instrumented steel H-piles accompanied with various in-situ
and laboratory soil tests were used. In the preliminary FE analysis, the soil constitutive
parameters were estimated, based on empirical correlations with Cone Penetration Test
(CPT) data. Two parameters were determined to represent the main potential sources for
error in the FE model. These two parameters were the soil-pile interface properties and the
soil elastic modulus. To improve the determination of these parameters, the soil-pile interface
properties were measured in the laboratory using a modified Direct Shear Test (mDST),
which directly measures the shear stress versus displacement relationship at soil-steelcontacts without the use of empirical correlations. In addition, the soil elastic modulus was
measured by means of a Consolidated Undrained (CU) Triaxial soil test. Measured soil
constitutive properties were used in the FE model, which showed significant improvement in
predicting the pile load-displacement behavior compared with the field measured responses
of the four load tested piles. A significant benefit of improving the simplified FE model is
that it can facilitate incorporation of the vertical settlement and serviceability limits into the
design of deep foundations.
6.3. Preliminary FE Model
For soil-structure interaction problems, the commercially-available finite element
code, PLAXIS, has been widely used and reported by many researchers to model the
behaviour of deep foundations with an acceptable accuracy (Prakoso and Kulhawy 2001;
Sellountou et al., 2005; Oh et al., 2008). In this study, PLAXIS was used because it is
capable of solving non- linear plasticity problems using a staged construction operation,
which allows for changing the load configuration on the model pile head, pore water pressuredistribution, and pile geometry. Ideally, a three-dimensional analysis is preferable to simulate
the behaviour of steel H-piles, but as iterated before, this is the first attempt to analyze steel
H-piles using a simplified axisymmetric model in sand and clay soils. The H-pile was
represented in the axisymmetric model assuming the cross-section of the actual pile is
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
181/235
164
circular. However, the model pile diameter was selected to provide the same surface area of
the actual H-pile along the shaft, while the cross-section of the model pile was gradually
reduced at the tip to reflect the actual area of the pile s tip. The H-pile was modeled using a
non-porous linear-elastic constitutive model with properties summarized in Table 6.1. To
eliminate the effects of the FE model vertical boundaries on the soil-pile response, the model
diameter was selected equal to 7.0D, where D is the model pile diameter. Also, a length equal
to 5D was defined between the model pile tip and the bottom horizontal boundary of the FE
model.
The preliminary FE analysis was conducted to model the behavior of four statically
load tested piles. Two of the test piles with identification numbers (ISU-4 and ISU-5) were
driven into cohesive soil layers and the other two test piles (ISU-9 and ISU-10) were driveninto mainly cohesionless soil layers. Table 6.1 summarizes soil layers and soil properties for
all test sites. As can be seen in the table, the unit weights of different soil layers were
measured in the laboratory, while the remaining soil constitutive parameters were estimated
using correlations with the CPT results. All the empirical correlations with the CPT data used
to estimate the soil properties are referenced in Table 6.1. In all cases, the Mohr-Coulomb
model was used to describe the constitutive soil behavior utilizing an unstructured very fine
mesh consisting of 15-node triangular elements. The Mohr-Coulomb elastic perfectly plastic
model was selected for this preliminary FE analysis because it is one of the most widely used
constitutive models and has proven to be quite accurate and efficient in simulating soil-
structure interaction problems (Tan and Bui, 2006). In addition to the elastic soil modulus
and Poissons ratio, E s and , respectively, the Mohr -Coulomb model incorporates the use of
three plastic parameters, , c, and , which denote the friction angle, cohesion, and dilatancy
angle of the soil material, respectively. The angle of dilatancy can be defined as the tendency
of soil to expand in volume due to shear. Table 6.1 includes all the required soil constitutive
parameters for this analysis based on the Mohr-Coulomb model.
In the preliminary FE models, the ground water table was defined for all test sites,
using the phreatic surface option of the program and the initial stresses were generated using
the predefined K o values based on the friction angle of the soil. For the interface element
between the soil and the pile, the PLAXIS code uses a virtual zero thickness element with
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
182/235
165
reduced properties that connect pile elements with surrounding soil elements. The reduced
interface properties are defined for each soil layer by determining the strength reduction
factor for interfaces (R inter ). This factor relates the strength of the interface (in terms of
friction angle, , and adhesion, a) to the soil cohesion and frictio n angle. Based on the
PLAXIS user manual recommendations, the R inter can be assumed equal to 0.50 and 0.67 for
clay-steel and sand-steel contacts, respectively. Table 6.1 summarizes the used values of the
R inter in the preliminary FE analysis for different soil layers presented at the four test sites.
Table 6.1: Constitutive soil parameters estimated using empirical correlations to CPT
TestSite
SoilLayers (1)
Thicknessm
drykN/m 3
wetkN/m 3
Su (2)kPa
c (3)kPa
(4) (5) E s(6)
kPa R inter
ISU-4(GWT:4.5m)
Clay 0 - 1.5 17 21.3 101 50 34 0.3 9347 0.5Clay 1.5 - 5.5 18.2 21.3 102 51 29 0.25 10586 0.5Clay 5.5 - 11 18.9 21.3 177 88 33 0.35 13662 0.5Clay 11 - 18 19.5 22.4 129 64 31 0.35 10366 0.5
ISU-5(GWT:10.0m)
Clay 0 - 7.6 17.5 21.1 107 54 32 0.30 9753 0.5Clay 7.6 - 12 19.5 22.9 148 74 32 0.30 13638 0.5Clay 12 - 18 20.5 23.8 180 90 34 0.35 16464 0.5
ISU-9(GWT:5.2 m)
Clay 0 - 4.8 18.8 23.4 216 108 34 0.25 17070 0.5Sand 4.8 - 13 18.5 22.9 0 0 40 0.45 28560 0.67Sand 13- 16 21.4 25.2 0 0 44 0.45 19490 0.67
ISU-10(GWT:3.0 m)
Sand 0 - 4.6 17.6 20.2 0 0 36 0.45 9248 0.67Sand 4.6 - 15 20.3 24.0 0 0 40 0.45 24390 0.67
Boulder 15 -17 24 24.0 0 0 44 (7) 0.25 270000 (7) 0.67For unit conversion: 1 m = 39.37 inch; 1 kN/m 3 = 6.24 lb/ft 3; and 1 kPa = 0.145 psi.(1) An undrained soil condition was used for the cohesive layers, while a drained condition was usedfor cohesionless layers; (2) After Schmertmann et al. (1978) using N k = 15;
(3) c = S u/2;(4) After
Robertson and Campanella (1983); (5) After Das (1998); (6) After Mayne and Kulhawy (1990) usingM=8.25 (q t- vo); and
(7)CPT not available, and E s were extrapolated. Pile: Linear-Elastic model;=50 kN/m/m; E s=2.0E8 kPa; =0.2; Shaft radius=0.16m; Tip radius=0.05m ; R inter =1.0.Steel Plate:Elastoplastic model; EA=1.6x10E8 kN/m; EI=9000 kN m2/m; d=0.026m; w=0.57 kN/m/m; =0.25.
6.3.1. Piles in cohesive soils
The two test piles referred to as ISU-4 and ISU-5 were selected for the preliminary
FE analysis in cohesive soils. Figure 6.1a shows different soil layers, boundary conditions,
and the interface elements used along the pile length for the preliminary FE model
representing the test pile ISU-4 (FE-ISU4-1 model). As can be seen from the figure, the
bottom boundary was totally fixed in both directions, while the side boundaries were only
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
183/235
166
fixed in the horizontal direction. The figure also shows the positive interface elements
connecting the soil and pile elements along different soil layers. The positive interface means
the interface is connecting the outer surface of the pile with the soil. Figure 6.1b shows a
view of finite element mesh generated for this analysis.
Figure 6.1: FE model representing the test pile at ISU-4
The FE analysis was performed based on controlled vertical displacement increments
that represented the same displacement values collected during the static load test on the test
piles. Using the staged construction option of the program, the displacements and stresses
were reset for each loading increment to avoid convergence issues that could take place in
the analysis. For the FE-ISU4-1 model, Figures 6.2a and 6.2b, respectively, show the
deformed mesh and the distribution of the total displacement corresponding to an extreme
prescribed vertical displacement of 40 mm (1.6 inch) at the model s pile head. As can be seen
from the figure, the maximum total displacement around the model s pile was limited to
around 2.5D, where D is the model pile diameter; hence, the FE model boundaries had noeffect on total displacement. Moreover, Figure 6.2c shows the plastic zone by means of the
effective normal stresses developed in the soil medium at the extreme total displacement. As
can be seen from the figure, the maximum effective normal stress induced in the soil is 410.6
kPa (59.6 psi), concentrated at an approximate distance of around 3D below the pile tip and
degrading towards the bottom horizontal boundary. Thus, the bottom boundary has a
(a) Model overview (b) Fine unstructured mesh
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
184/235
167
negligible effect on the stresses induced in the soil below the pile tip.
Figures 6.3a and 6.3b show the predicted load-displacement response at the pile head
compared to the measured response for the test piles ISU-4 and ISU-5, respectively. From
Figure 6.3a, it can be seen the predicted load-displacement curve is less stiff by about 60%
compared to the measured response along the elastic portion. Moreover, the predicted load-
displacement curve underestimated the pile nominal capacity at the plastic portion by 43%
when compared with the measured pile nominal capacity. For each case, the measured pile
nominal capacity was defined using the Davisson criterion (Davisson, 1972) represented in
the figures by a dot line referred to as the Davisson line. For the FE model that represented
the test pile ISU-5 (FE-ISU5-1 model), Figure 6.3b shows the predicted load-displacement
curve has stiffness 64% less than that of the measured curve along the elastic portion, and65% lower nominal capacity compared to the measured capacity. Thus, the preliminary FE
analyses for vertically-loaded piles driven into cohesive soils generally show poor
correlations compared to field measured responses. The previous conclusion was expected,
as several studies reported estimated soil-pile interface properties and soil elastic modulus,
based on empirical correlations that could lead to biased FE results (El-Mossallamy 1999;
De-Gennaro 2006; Engin 2007).
Figure 6.2: FE model results for the test pile at ISU-4
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
185/235
168
Figure 6.3: Predicted versus measured pile load-displacement responses for piles in clay
6.3.2. Piles in cohesionless soils
The same axisymmetric FE model was utilized to simulate the behavior of the other
two test piles driven into mainly cohesionless soil layers and referred to as ISU-9 and ISU-
10. However, the top soil layer at ISU-9 was classified as low plastic clay with thickness
equal to 4.8m (15.74 ft). As summarized in Table 6.1, all the constitutive parameters were
established for the soil material, based on empirical correlations to the CPT conducted at
each test site. This table also summarizes the selected R inter values for each soil layer, based
on the recommended values in the PLAXIS user manual. Figure 6.4a shows different soillayers, boundary conditions, and the interface elements used along the pile length for the
preliminary FE model representing ISU-9 (FE-ISU9-1 model). Moreover, Figure 6.4b shows
a view of finite element mesh used for the analysis.
For the FE-ISU9-1 model, Figures 6.5a and 6.5b, respectively, show the deformed
mesh and the distribution of the total displacement corresponding to the extreme prescribed
vertical displacement of 18.4 mm (0.72 inch) at the model s pile head. As can be seen from
Figure 6.5a, the maximum total displacement around the model s pile was concentrated in
the top clay layer and below the model s pile tip. However, the displacement did not extend
laterally more than around 5.0D. Hence, the FE model boundaries had minimal effects on
total displacements. Moreover, Figure 6.5c shows the plastic zone by means of effective
normal stresses developed in the soil medium at the extreme total displacement. As seen in
this figure, the maximum effective stress induced in the soil is 421.9 kPa (61.2 psi), which is
0 20 40 60 80 100 120 140 160 180
0
0.2
0.4
0.60.8
1
1.2
1.4
1.6
1.8
0
5
10
15
20
25
30
0 100 200 300 400 500 600 700 800
Load (kips)
D i s p
l a c e m e n t ( i n c h
)
D i s p
l a c e m e n t ( m
m )
Load (kN)
SLT (ISU4)FE-ISU4-1
0 40 80 120 160 200 240 280
0
0.2
0.4
0.60.8
1
1.2
1.4
1.6
1.8
0
5
10
15
20
25
30
35
0 200 400 600 800 1000 1200
Load (kips)
D i s p
l a c e m e n t ( i n c h
)
D i s p
l a c e m e n t ( m
m )
Load (kN)
SLT (ISU5)
FE-ISU5-1
(b) Test pile at ISU-5 (clay)(a) Test pile at ISU-4 (clay)
-
7/27/2019 Behavior Characterization and Development of LRFD Resistance Fact
186/235
169
concentrated directly below the model pile tip and degrading towards the bottom horizontal
boundary.
Figure 6.4: FE model representing the test pile at ISU-9
Figure 6.5: FE model results for the test pile at ISU-4
From Figure 6.6a, it can be seen that the predicted load-displacement curve